Significance of profiles in direct imaging process.

Agic, Darko ; Strgar Kurecic, Maja ; Gojo, Miroslav 等

Abstract: Profiles are today very important tool in graphic reproduction process. They describe and characterise possible reproductive colors that an input and/or autput device may achieve in selected and defined conditions. For various reasons indirect capturing way was often used, so color transparency or print was used as an original for imaging. This procedure not only elongates the reproduction process, but can significantly cause serious disturbances and shifts in tonal as well as color reproduction, particularly at specific and sensitive originals with non standard colorants. The aim of this research was to make a color quality comparison between standard and direct imaging way (workflow) using dedicated profiles for reproducing originals characterised with non common colors.

Key words: profiles, reproduction procedure, non standard colours

1. INTRODUCTION

In recent reproduction process efficiency and quality of reproduction can significantly increase by implementing colour management system (CMS) already at the beginning of the reproduction chain. Recent procedures of some originals or scenes were captured (photographed) on standard colour material, so transparency or colour print acted as an secondary original, fig 1. There were limitations in light sources (colour temperature of sources had to be adjusted to colour materials, daylight or studio illumination, ISO 12646), and colour film itself as well as chemical process of developing could also produce deviations in reproduction of colours. In such situation the profile (input profile for camera and colour film) was not an optimal solution, because it had to be linked with next stage, the profile of the digitalisation device e.g. scanner. The CMS coordinates and transforms these information for other stages, leading to the output devices. In direct imaging process the appropriate digital photographic system could be used, so colour film and chemical processes can be avoided. That not only shortens and condenses the process, but yields enhanced quality, because input profile is simpler to be achieved with no possible limitations of the interstate. For calibrating and profiling reproduction devices various targets as tests are used. Usually, most of them are "all purpose", containing various number of coloured patches distinctly arranged (although characterized in ISO 12641, 1997). In the situation of required reproduction of some special originals such as art paintings made with non standard colorants, meaning non standard gamut and colour specifications (Mortimer 1998), a custom designed input target supporting such demands, has to be provided. This work covers a situation of reproducing some non standard original, gouache painting, on both ways, recent and direct imaging, and colorimetric comparison of some reproduced colours. Profile for selected output CMYK device (printer) was also established.

2. BACKGROUND

The efficiency and quality during the reproduction can be significantly influenced by implementation of CMS at the beginning of the process. As the photographic system is the input device in the input/output reproduction system, its ability of precise interpretation of the imaged scene significantly contributes exact interpretation of colours, and entire quality of reproduction. That includes precise calibration and characterisation of the system.

[FIGURE 1 OMITTED]

Characterisation process includes profile generation, so called profiling. The profile describes device behaviour, and gives information about colours that device is able to generate, respectively colours gamut that the device can obtain. Characterisation of the device includes various stages: from the selected test target known RGB or CMYK values are sent, generated and colorimetric measured. That is followed by a profile accomplishing application that interconnects RGB and CMYK values of colours achieved by the device with their equivalents in referent colour CIE XYZ or Lab space. According to colorimetric values profile notifies the CM module about the content of RGB or CMYK amount to represent the referred colour, and Colour management system what objective colour will result from the given RGB or CMYK values set. Although both processes are similar, characterisation of a scanner formally appears to be simpler as it uses a fixed light source, and scanning originals are mainly CMY or CMYK based (usually slides or colour prints, real originals are directly processed rarely). On the other hand it covers more stages than can bring deviations and/or some limitations in the process. For direct imaging if used in nature surrounding where illumination is mixed or varying, it is very difficult to achieve stabile and persistent characterisation of the system. In studio environment with defined and stabile conditions, specially illumination, but in special situations where other surrounding parameters could be present too, advantages of direct imaging photographic system are present. Characterisation is possible, particularly for situation of imaging non standard or specific originals containing non standard colorants, wide density range, large gamut or shift in gamut, as well some other specific optical or visual characteristic or behaviour (Berns R.S. 2000). Special problem might appear at some artworks light or heat sensitive, that have to be treated with special care, as they could obtain cumulative effect of acting influences. As the output device generally is CMYK (or extended CMYK) based, it is desired to be a quality one with appropriate profile, powered by suitable RIP device, for obtaining optimal possible colour gamut and control of the screening system. While tuning the image file for output it is very important to recheck selected parameters for profiles, illumination, achromatic functions, gain, rendering intent, printing system etc to avoid possible reproduction confusions.

3. EXPERIMENTAL

In former studies it was decided what illumination conditions should be suitable (3200 K and 5500 K) for both imaging processes. Spectral analysis of selected colours of specified artwork was achieved, so that custom target could be prepared. In this step it is important to prepare patches whose optical and spectral characteristics correspond to these on selected kind of artwork original. In that stage 24 patches were observed, and their values were applied to profiling software (Gretag-Macbeth Profile maker). Imaging was performed, for comparison, in recent photographic system and then with direct digital image capturing. In both processes profiling target was applied. Characterisation of the chosen colour printer, as the output device, was made too.

[FIGURE 2 OMITTED]

4. RESULTS AND DISCUSSION

Comparing results from recent and direct imaging process, colorimetric differences are some smaller in direct imaging process. (table 1.) The mainly reason for colorimetric difference in recent system used is not saturation change, than lightness and tone decrease, than in direct process. Tone deviations are mostly bounded with imaging illumination conditions, and minor deviations are present with profiles with artificial light with. Imaging with artificial light without correction filter are more unsuitable, as there significant tone changes and saturation loss is achieved. It has to be mentioned that recent system using transparency or colour print can obtain good results, but they can considerably depend on colour material characteristics and scanner quality. Implementing of colour management in reproduction process enables more efficient bounding various devices. Profile implementation of input and output devices of the reproduction process improves the colour interpretation and accuracy (Fig 2).

Properly defined output device will not achieve good results if input device profile characterisation fails. Although, custom test target if applied, containing colour patches with optical if input device profile characterisation fails. Although, custom test target if applied, containing colour patches with optical properties similar to original can achieve a profile that better interprets original colours, than using a standard all purpose chart. Besides enhanced possible quality of reproduction using colour management with custom profiling, a abbreviation of process time, editing of picture, is obtained, disadvantages of colour materials are skipped, and advantages of direct imaging system stay present.

5. REFERENCES

ISO 17321 WDH4: Colour characterisation of digital still cameras using colour targets and spectral illumination, ISO 12646: Graphic technology--Displays for color proofing -characteristics and viewing conditions, ISO 13655 Graphic technology: Spectral measurements and colorimetric computation for graphic arts images.

ISO 12641:Graphic technology-Prepress digital exchange-color targets for input scanner calibration

Mortimer A., (1998) Color reproduction in a digital age; PIRA International, UK.

Berns R.S., (2000) The Science of Digitizing Two Dimensional Works of--Art for Color-Accurate Image Archives, Munsell Color Science Laboratory Technical Report

Table 1: Calculated average colorimetric differences L, C and
H, and colorimetric differences [DELTA][E.sub.94] and
[DELTA][E.sub.2000] for a selected original

 Original [DELTA]L [DELTA]C [DELTA]H
 painting Avg. Avg. Avg.

3200 K Custom 1,00 -2,85 5,36
 target

 Standard -0,04 -4,07 5,00
 target 1

 Without 7,59 -7,72 7,57
 profile T

 Analogue -3,62 -10,61 7,08
 imaging
 process

5500 K Custom -0,62 -2,80 2,41
 target

 Standard -2,66 -6,27 1,90
 target 1

 Without 5,63 -12,32 2,36
 profile D

 Analogue -5,89 -6,55 6,10
 imaging
 process

 Original [DELTA][E.sub.94] [DELTA][E.sub.00]
 painting Avg. Avg.

3200 K Custom 5,08 5,56
 target

 Standard 5,64 6,10
 target 1

 Without 11,78 10,87
 profile T

 Analogue 10,00 9,85
 imaging
 process

5500 K Custom 3,24 3,09
 target

 Standard 4,61 4,11
 target 1

 Without 9,10 8,46
 profile D

 Analogue 8,84 9,85
 imaging
 process